A great need still exists for new therapies to treat tumor progression and to prevent tumor metastasis. The matrix metalloproteinases (MMPs) are one family of enzymes that have a clear role in these events. The long-term objective of the proposed study is to derive an understanding of the structure and function of the MMP active site and to apply this understanding to bring forward new antagonists of the MMPs as anti-tumor agents. An overwhelming body of evidence indicates that matrix metalloproteinases (MMPs) are causally involved in tumor progression and metastasis. Yet a decade-long effort of developing matrix metalloproteinase inhibitors (MMPIs) and testing them in the clinical arena has now lost momentum. This creates a paradox;we know the MMPs are valid targets, but the drugs were disappointing in clinical trials. A number of factors contributed to the failure of the MPIs in the clinic. These factors include an incomplete understanding of the pathophysiology of MMPs in cancer, the inability to monitor MMP inhibition in vivo, and the lack of selective inhibitors. The objective of the proposed study is to overcome the hurdles that prevented the MMPI's from realizing their promise in clinical trials in oncology. The Specific Aims of the proposed study are to: 1) Identify optimal and selective peptide substrates for tumor-associated MMPs and then use these to develop fluorescent, near-infrared fluorescent, and magnetic resonance imaging agents that report on the MMP activity in vivo. The substrates will also be used as a guide to probe structure-activity relationships within the MMP active site. 2) Develop a publicly accessible computational environment, called Proteolysis MAP, to guide reasoning about proteolysis on a genome scale. 3) Use mass spectrometry to identify the pathophysiologic substrates for MMP-7, MMP-2 and MMP-9, with particular emphasis on mouse models where these MMPs are known to be involved in tumor progression. 3) Design novel and selective small molecule antagonists of MMP-7 and MMP-9. Virtual docking and/or NMR screening will be used to identify chemical leads that hit the 82 subsite in the catalytic cleft. These chemical leads will be improved by iterative structure-based modifications and will be tested in cell-based assays and ultimately in animal models of tumor progression.